A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio
The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2018 |
---|---|
Erschienen: |
2018 |
Enthalten in: |
Zur Gesamtaufnahme - volume:40 |
---|---|
Enthalten in: |
IEEE transactions on pattern analysis and machine intelligence - 40(2018), 8 vom: 20. Aug., Seite 2023-2029 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Sourati, Jamshid [VerfasserIn] |
---|
Links: |
---|
Themen: |
Journal Article |
---|
Anmerkungen: |
Date Completed 29.08.2019 Date Revised 01.02.2022 published: Print-Electronic Citation Status MEDLINE |
---|
doi: |
10.1109/TPAMI.2017.2743707 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM275310647 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM275310647 | ||
003 | DE-627 | ||
005 | 20231225005405.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2018 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1109/TPAMI.2017.2743707 |2 doi | |
028 | 5 | 2 | |a pubmed24n0917.xml |
035 | |a (DE-627)NLM275310647 | ||
035 | |a (NLM)28858784 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Sourati, Jamshid |e verfasserin |4 aut | |
245 | 1 | 2 | |a A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio |
264 | 1 | |c 2018 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 29.08.2019 | ||
500 | |a Date Revised 01.02.2022 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status MEDLINE | ||
520 | |a The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, N.I.H., Extramural | |
650 | 4 | |a Research Support, U.S. Gov't, Non-P.H.S. | |
700 | 1 | |a Akcakaya, Murat |e verfasserin |4 aut | |
700 | 1 | |a Erdogmus, Deniz |e verfasserin |4 aut | |
700 | 1 | |a Leen, Todd K |e verfasserin |4 aut | |
700 | 1 | |a Dy, Jennifer G |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t IEEE transactions on pattern analysis and machine intelligence |d 1979 |g 40(2018), 8 vom: 20. Aug., Seite 2023-2029 |w (DE-627)NLM098212257 |x 1939-3539 |7 nnns |
773 | 1 | 8 | |g volume:40 |g year:2018 |g number:8 |g day:20 |g month:08 |g pages:2023-2029 |
856 | 4 | 0 | |u http://dx.doi.org/10.1109/TPAMI.2017.2743707 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 40 |j 2018 |e 8 |b 20 |c 08 |h 2023-2029 |